Forty male volunteers were randomly assigned to one of four treatment groups on a double-blind basis: (1) Imipramine--25 mg t.d.s., (2) Viloxazine--50 mg t.d.s., (3) Placebo, and (4) Control--no tablets. Tests were carried out (1) before treatment, (2) 2 h after the first dose, (3) on Day 3 after 7 doses, and (4) on Day 7 after 21 doses. The driving tasks consisted of (1) weaving around a series of bollards while simultaneously responding to an auditory logic task and (2) a gap acceptance task. Using an analysis of covariance repeated measures design, it was found that imipramine tended to increase the level of risk acceptable to the subject as compared to either placebo or control. Imipramine also impaired performance on other tasks. Viloxazine appeared to be little different from either placebo or control on any of the tasks.